Abstract

Cesare Cozzarini¹, Giuseppe Sanguineti², Vittorio Vavassori³, Fernando Munoz⁴, Barbara Avuzzi⁵, Elisabetta Garibaldi⁶, Domenico Cante⁷, Justina Magdalena Waskiewicz⁸, Alessandro Magli⁹, Adriana Faiella¹⁰, Elisa Villa³, Letizia Ferella⁴, Marco Gatti¹¹, Barbara Noris Chiorda⁵, Cristina Piva⁷, Paolo Ferrari¹², Tiziana Rancati¹³, Fabio Badenchini¹³, Giuseppe Girelli¹⁴, Eugenia Moretti¹⁵, Riccardo Valdagni¹⁶, Andrea Bresolin¹⁷, Nadia Di Muzio¹⁸, Claudio Fiorino¹⁷

¹San Raffaele Scientific Institute, Radiotherapy, Milan, Italy; ²IRCCS Istituto Nazionale dei Tumori “Regina Elena", Radiotherapy, Rome, Italy; ³Cliniche Gavazzeni Humanitas, Radiotherapy, Bergamo, Italy; ⁴Ospedale Regionale Parini-AUSL Valle d’Aosta, Radiotherapy, Aosta, Italy; ⁵Fondazione IRCCS Istituto Nazionale dei Tumori, Radiotherapy, Milan, Italy; ⁶A.O. SS. Antonio e Biagio, Radiotherapy, Radiotherapy, Alessandria, Italy; ⁷Ospedale di Ivrea, Radiotherapy, Ivrea, Italy; ⁸Azienda Sanitaria dell’Alto Adige, Radiotherapy, Bolzando, Italy; ⁹Azienda Ospedaliero Universitaria S. Maria della Misericordia, , Radiotherapy, Udine, Italy; ¹⁰IRCCS Istituto Nazionale dei Tumori “Regina Elena”, Radiotherapy, Rome, Italy; ¹¹IRCC Candiolo, Radiotherapy, Candiolo, Italy; ¹²Azienda Sanitaria dell’Alto Adige, Radiotherapy, Bolzano, Italy; ¹³Fondazione IRCCS Istituto Nazionale dei Tumori - Prostate Program, Radiotherapy, Milan, Italy; ¹⁴Ospedale degli Infermi, Radiotherapy, Biella, Italy; ¹⁵Azienda Ospedaliero Universitaria S. Maria della Misericordia, Radiotherapy, Udine, Italy; ¹⁶Fondazione IRCCS Istituto Nazionale dei Tumori, Prostate Program - Università degli Studi Milano, Radiotherapy, Milan, Italy; ¹⁷San Raffaele Scientific Institute, Medical Physics, Milan, Italy; ¹⁸Università Vita Salute San Raffaele, Radiotherapy, Milan, Italy

To identify predictors of patient-reported urinary incontinence (PRUI) two years after radical, adjuvant or salvage  (RAD, ADV, SALV) radiotherapy for prostate cancer (PCa).

A prospective cohort of 407 men enrolled in a multi-Institutional, observational registered study investigating radiation-induced toxicity, with complete ICIQ-SF questionnaire at baseline and 2 years after RT start, was analyzed. Patient characteristics: RAD n=141, cT1/T2/T3a/T3b 19/37/33/11%, median 2-Gy equivalent dose (mEQD2) 80 Gy; ADV n=104, pT2/pT3a/pT3b 8/36/56%, mEQD2 70.41 Gy, median time to RT (mTTRT) from prostatectomy 3.9 months; SALV n=162, pT2/pT3a/pT3b 45/33/22%, mEQD2 74 Gy, mTTRT 19.1 months. Androgen deprivation was given to 95%, 56% and 37% of pts treated with RAD, ADV and SALV intent, respectively, for a median of 24 months. Frequency and amount of urine leakage and their impact on self-perceived QoL (ICIQ-SF3, 4 and 5, respectively) were analyzed, as well as the overall ICIQTOT score. The 75^th percentiles of ICIQ-SF3, 4, 5 and TOTAL worsening at 2 years with respect to the baseline were considered as end-points. Severe PRUI (ICIQTOT≥13 points) at two years was also analyzed. The association between these end-points and many clinical and dosimetric variables, including 2-Gy equivalent RT doses (EQD2), RT intent, use of concomitant ADT, comorbidities, patient personality as measured by the EPQR questionnaire, baseline Bowel, Emotional, Social and Systemic Domains as evaluated by the IBDQ questionnaire, was investigated by univariable (UVA) analyses. Variables with a è-value <0.20 at UVA were entered into a multivariable backward logistic regression analysis (MVA).

No role emerged (even at UVA) for EQD2. Only baseline ICIQTOT (OR 1.32, p>0.0001) and SALV intent (OR 3.143, p=0.004) independently predicted an ICIQTOT score at 2 years ≥13 points. A summary of the predictors of PRUI worsening is shown in Table 1, including the performance of the resulting models in terms of area under curve (AUC) and goodness of fit (Hosmer-Lemeshow test).

Baseline urinary incontinence status and radiotherapy intent are the major factors predicting 2-year patient-reported urinary incontinence. The impact of RT intent (ADV/SALV vs RAD) is stronger for the “objective” (frequency and amount) PRUI symptoms. Both psychoticism and lie personality trait (a measure of social acquiescence) significantly reduce patient perception of the detrimental effect of PRUI on their QoL.